Biology is never simple. While researchers make progress in reading and publishing genes to treat the disease, for example, an increasing set of evidence suggests that proteins and metabolites surrounding these genes cannot be ignored.
MIT Spinout ReviveMed has created a platform to measure metabolites – metabolism products such as lipids, cholesterol, sugar and carbohydrates – on a large scale. The company uses these measures to find out why some patients react to treatments when others do not do so and to better understand the engines of the disease.
“Historically, we were able to measure a few hundred metabolites with great precision, but it is a fraction of the metabolites that exist in our body,” explains the CEO of ReviveMed Leila Pirhaji Phd '16, who founded the company with Professor Ernest Fraenkel. “There is a massive gap between what we measure with precision and what exists in our body, and that is what we want to approach. We want to exploit the powerful ideas of underused metabolite data. ”
Revivemed’s progress occurs while the wider medical community is increasingly linking metabolites, deregulated to diseases such as cancer, Alzheimer's disease and cardiovascular disease. ReviveMed uses its platform to help some of the largest pharmaceutical companies in the world to find patients who benefit from their treatments. It also offers software to university researchers for free to help get information from unusual metabolite data.
“With the booming AI field, we think we can overcome the data problems that have limited the study of metabolites,” explains Pirhaji. “There is no basic model for metabolomics, but we see how these models modify various fields such as genomics, we therefore start to pioneer of their development.”
Find a challenge
Pirhaji was born and grew up in Iran before coming to MIT in 2010 to continue his doctorate in biological engineering. She had already read Fraenkel's research articles and was delighted to contribute to the network models he built, which integrated data from sources such as genomes, proteome and other molecules.
“We thought of the overall view in terms of what you can do when you can measure everything – genes, RNA, proteins and small molecules such as metabolites and lipids,” explains Fraenkel, who is currently sitting on the board of directors of ReviveMed. “We are probably able to measure something only 0.1% of small molecules in the body. We thought there must have been a way to get such a complete view of these molecules as we did for others. This would allow us to draw all the changes in the cell, whether in the context of cancer or development or development or degeneration. ”
Halfway through his doctorate, Pirhaji sent some samples to a collaborator at Harvard University to collect data on the metabolome – small molecules which are the products of metabolic processes. The employee returned Pirhaji a huge Excel sheet with thousands of data data – but they told him that she was better to ignore everything beyond the best lines because they did not know what other data meant. She took this as a challenge.
“I started to think that we might be able to use our network models to solve this problem,” recalls Pirhaji. “There was a lot of ambiguity in the data, and it was very interesting for me because no one had tried this before. It seemed to be a big gap in the field.”
Pirhaji has developed a huge graph of knowledge that included millions of interactions between proteins and metabolites. The data were rich but messy – Pirhaji described it as “hair ball” that could not say anything to the researchers on the disease. To make it more useful, it has created a new way of characterizing metabolic routes and features. In a 2016 article in Nature methodsShe described the system and used it to analyze metabolic changes in a Huntington disease model.
Initially, Pirhaji had no intention of starting a business, but it began to carry out the commercial potential of technology in the last years of its doctorate.
“There is no entrepreneurial culture in Iran,” explains Pirhaji. “I did not know how to start a business or transform science into startup, so I exploited everything that MIT offered.”
Pirhaji began to take courses at MIT Sloan School of Management, including Cours 15.371 (innovation teams), where she has teamed up with classmates to think about how to apply her technology. She also used the MIT Venture mentoring service and the Sandbox MIT, and participated in the Accelerator of Startup Delta V of Martin Trust Center for MIT Entrepreneurship.
When Pirhaji and Fraenkel officially founded ReviveMed, they worked with the office of the granting of MIT technological licenses to access the patents around their work. Pirhaji has since developed the platform to solve other problems which she discovered during talks with hundreds of managers in pharmaceutical companies.
ReviveMed started by working with hospitals to discover how lipids are deregulated in a disease known as the steathepatitis associated with metabolic dysfunction. In 2020, ReviveMed worked with Bristol Myers Squibb to predict how the subsets of cancer patients would react to the company's immunotherapies.
Since then, ReviveMed has worked with several companies, including four of the 10 best world pharmaceutical companies, to help them understand the metabolic mechanisms behind their treatments. These ideas help identify patients who benefit the most from different therapies.
“If we know which patients will benefit from each medication, it would really decrease the complexity and time associated with clinical trials,” explains Pirhaji. “Patients will get good treatments faster.”
Generative metabolomics models
Earlier this year, ReviveMed has collected a set of data based on 20,000 patients' blood samples he used to create digital patient twins and generative AI models for metabolomic research. ReviveMed puts its generative models available to non -profit university researchers, which could accelerate our understanding of how metabolites influence a range of diseases.
“We democratize the use of metabolomic data,” explains Pirhaji. “It is impossible for us to have data from each patient in the world, but our digital twins can be used to find patients who could benefit from treatments according to their demography, for example, by finding patients who could be at risk of cardiovascular disease.”
The work is part of ReviveMed's mission to create metabolic foundation models that researchers and pharmaceutical societies can use to understand how diseases and treatments modify patient metabolites.
“Leila has solved a lot of really difficult problems you face when you try to remove an idea from the laboratory and transform it into something robust and reproducible to be deployed in biomedicine,” says Fraenkel. “Along the way, she also realized that the software she has developed is incredibly powerful by himself and could be transformational.”
